I’ve sat on both sides of this table.
As a designer, I’ve had pieces come back with the edge softened, the wall thickened, the finish swapped for an “equivalent” — every change defensible, the design gone. As a production manager, I’ve been the person making exactly those changes, for reasons that were completely sound from where I stood.
So believe me: the factory isn’t wrong, and neither are you. You’re speaking two different languages about the same object, and almost nothing translates itself.
When a design goes into production, the obvious things survive. Silhouette, dimensions. What erodes is finer: proportion, edge, weight, how light sits on a surface, the restraint that made the piece feel inevitable. None of that is written in your drawing, and no workshop can infer it from one.
The factory’s list is real: can it be made, how, at what cost, what yield, will the process hold. Your list is real too: does the material still have presence, did a detail get simplified into coarseness, is the thing still worth making at all. Two precise lists. Almost no shared vocabulary.
Left alone, the gap fills with defaults. A radius grows because the tooling prefers it. A wall thickens because the caster is being safe. “Polished” means one thing in your render and another on the workshop floor. Each change is small, professional, defensible. Three of them together, and the piece that arrives is recognisably yours — and not yours at all.
Why more communication hits a ceiling
The standard answer is to communicate harder: more photos, more calls, more revisions. That helps — up to a sharp limit. Communication can only carry what you know needs saying. The changes that kill a design come from risks you didn’t know existed. You never wrote “this edge is the whole design,” because it never occurred to you that anyone would touch the edge. You can’t flag a danger you can’t see, and you can’t see it unless you know what the process wants to do to your piece before it does it.
The other obvious answer: shouldn’t a good factory do this translation itself? Some genuinely try. Workshops with real design literacy exist, and finding one is worth a great deal. But even the best workshop resolves ambiguity toward its own constraints — yield, schedule, process stability — because that’s what it’s accountable for. Not cynicism; its job. When your drawing is silent, the factory answers with its defaults. Translation has to sit with someone whose accountability is the design.
Translation is a judgment, not a relay
So what protects a design is someone who speaks both languages and decides, not forwards. Where cost should be spent. Where complexity should be removed. Which factory preference strengthens the design, and which one quietly kills it.
If you’re doing this yourself, four habits help. Name the three dimensions on your drawing that are sacred — in writing. Price the design twice, as drawn and as the workshop would prefer it, so the differences become choices instead of surprises. Ask the factory what they would change before they change it. And when a cost saving is about to land on the one detail carrying the whole piece, move it somewhere invisible.
And be honest about whether you need this layer at all, because it isn’t free. The maths is simple: the more of your design lives in the fine qualities — edge, weight, surface, proportion — the more translation is worth. If the piece survives a millimetre of drift and a substituted finish, skip it; you’re buying a standard part, and the factory’s language alone will serve you fine. But if the design is the kind that dies from three defensible changes, the translation layer isn’t overhead. It’s the difference between the object you drew and the object that shows up.
Most products don’t fail in design. They fail between the two languages. That space can be worked — it just can’t be worked by shouting across it.